UEF//eRepository
DSpace https://erepo.uef.fi
Rinnakkaistallenteet Terveystieteiden tiedekunta
2018
Body fat mass, lean body mass and associated biomarkers as
determinants of bone mineral density in children 6-8 years of age - The
Physical Activity and Nutrition in Children (PANIC) study
Soininen, Sonja
Elsevier BV
article
info:eu-repo/semantics/acceptedVersion
© Elsevier Inc.
CC BY-NC-ND https://creativecommons.org/licenses/by-nc-nd/4.0/
http://dx.doi.org/10.1016/j.bone.2018.01.003
https://erepo.uef.fi/handle/123456789/6125
Downloaded from University of Eastern Finland's eRepository
1
Body fat mass, lean body mass and associated biomarkers as
1
determinants of bone mineral density in children 6–8 years of age – The
2
Physical Activity and Nutrition in Children (PANIC) Study
3 4
Sonja Soininena,b,c,*, Virpi Sidoroffd, Virpi Lindia, Anitta Mahonene, Liisa Krögerf, Heikki 5
Krögerg,h, Jarmo Jääskeläinenf, Mustafa Atalaya, David E. Laaksonena,i, Tomi Laitinenj, Timo A.
6
Lakkaa,j,k 7
8
Affiliations:
9 10
a Institute of Biomedicine, Physiology, School of Medicine, University of Eastern Finland, PO Box 11
1627, 70211 Kuopio, Finland 12
b Institute of Dentistry, University of Eastern Finland, PO Box 1627, 70211 Kuopio, Finland 13
c Social and Health Center, City of Varkaus, Savontie 55, 78300 Varkaus, Finland 14
d Department of Pediatrics, North-Karelia Central Hospital, Tikkamäentie 16, 80210 Joensuu, 15
Finland 16
e Institute of Biomedicine, Medical Biochemistry, School of Medicine, University of Eastern 17
Finland, PO Box 1627, Kuopio, Finland 18
f Department of Pediatrics, Kuopio University Hospital and University of Eastern Finland, PO Box 19
100, 70029 Kuopio, Finland 20
g Department of Orthopedics and Traumatology, Kuopio University Hospital, PO Box 100, 70029 21
Kuopio, Finland.
22
h Kuopio Musculoskeletal Research Unit (KMRU), University of Eastern Finland, PO Box 1627, 23
70211 Kuopio, Finland.
24
i Department of Internal Medicine, Kuopio University Hospital, PO Box 100, 70029 Kuopio, 25
Finland 26
j Department of Clinical Physiology and Nuclear Medicine, Kuopio University Hospital, PO Box 27
100, 70029 Kuopio, Finland 28
k Kuopio Research Institute of Exercise Medicine, Haapaniementie 16, 70100 Kuopio, Finland 29
30
*Corresponding author: Sonja Soininen, University of Eastern Finland, Institute of Biomedicine / 31
Physiology, PO Box 1627, Fin-70211 Kuopio, Finland; sonja.soininen@uef.fi 32
33
e-mail addresses: sonja.soininen@uef.fi; virpi.sidoroff@siunsote.fi; virpi.lindi@uef.fi;
34
anitta.mahonen@uef.fi; liisa.kroger@kuh.fi; heikki.kroger@kuh.fi; jarmo.jaaskelainen@uef.fi;
35
mustafa.atalay@uef.fi;david.laaksonen@uef.fi; tomi.laitinen@kuh.fi; timo.lakka@uef.fi 36
2 Abstract
37
Lean body mass (LM) has been positively associated with bone mineral density (BMD) in children 38
and adolescents, but the relationship between body fat mass (FM) and BMD remains controversial.
39
Several biomarkers secreted by adipose tissue, skeletal muscle, or bone may affect bone metabolism 40
and BMD. We investigated the associations of LM, FM, and such biomarkers with BMD in children.
41 42
We studied a population sample of 472 prepubertal Finnish children (227 girls, 245 boys) aged 6-8 43
years. We assessed BMD, LM, and FM using whole-body dual-energy x-ray absorptiometry and 44
analysed several biomarkers from fasting blood samples. We studied the associations of LM, FM, 45
and the biomarkers with BMD of the whole body excluding the head using linear regression analysis.
46 47
LM (standardized regression coefficient β=0.708, p<0.001), FM (β=0.358, p<0.001), and irisin 48
(β=0.079, p=0.048) were positive correlates for BMD adjusted for age, sex, and height in all children.
49
These associations remained statistically significant after further adjustment for LM or FM. The 50
positive associations of dehydroepiandrosterone sulphate (DHEAS), insulin, homeostatic model 51
assessment for insulin resistance (HOMA-IR), leptin, free leptin index, and high-sensitivity C- 52
reactive protein and the negative association of leptin receptor with BMD were explained by FM. The 53
positive associations of DHEAS and HOMA-IR with BMD were also explained by LM. Serum 25- 54
hydroxyvitamin D was a positive correlate for BMD adjusted for age, sex, and height and after further 55
adjustment for FM but not for LM. LM and FM were positive correlates for BMD also in girls and 56
boys separately. In girls, insulin, HOMA-IR, leptin, and free leptin index were positively and leptin 57
receptor was negatively associated with BMD adjusted for age, height, and LM. After adjustment for 58
age, height, and FM, none of the biomarkers was associated with BMD. In boys, leptin and free leptin 59
index were positively and leptin receptor was negatively associated with BMD adjusted for age, 60
height, and LM. After adjustment for age, height and FM, 25(OH)D was positively and IGF-1 and 61
leptin were negatively associated with BMD. FM strongly modified the association between leptin 62
and BMD.
63 64
LM but also FM were strong, independent positive correlates for BMD in all children, girls, and boys.
65
Irisin was positively and independently associated with BMD in all children. The associations of other 66
biomarkers with BMD were explained by LM or FM.
67
3 Keywords: bone mineral density; lean body mass; body fat mass; DXA; child; cytokine
68
Abbreviations 69
BF%, body fat percentage 70
BMC, bone mineral content 71
BMD, bone mineral density 72
BMI, body mass index 73
DHEAS, dehydroepiandrosterone sulphate 74
DXA, dual-energy x-ray absorptiometry 75
FM, body fat mass 76
HOMA-IR, the homeostatic model assessment for insulin resistance 77
hs-CRP, high-sensitivity C-reactive protein 78
IGF-1, insulin-like growth factor 1 79
IL-6, interleukin 6 80
LM, lean body mass 81
SD, standard deviation 82
SDS, standard deviation score 83
TNF-α, tumor necrosis factor α 84
25(OH)D, 25-hydroxyvitamin D 85
4 1. Introduction
86
Early childhood and puberty are the periods of rapid growth and bone accretion, and the majority of 87
bone mass is gained during adolescence and early adulthood [1–3]. Bone mineral accrual during 88
growth is dependent on multiple factors such as genetic background, sex, race, nutrition, physical 89
activity, and hormone metabolism [2,3]. Higher lean body mass (LM) has been associated with higher 90
bone mineral density (BMD) and bone mineral content (BMC) in children and adolescents [4–7], but 91
the relationship of body fat mass (FM) with BMD or BMC remains controversial [5,6,8–10]. FM has 92
been positively associated with BMD independent of LM in prepubertal children [6]. However, there 93
is some evidence that higher FM is detrimental to bone accrual during and after puberty [5,8,9] and 94
that overweight children and adolescents are at an increased risk of forearm fractures [10].
95 96
Mechanical loading increases bone formation, and weight-bearing exercise improves bone mineral 97
accrual [11]. The classical Wolff’s law and later the Frost’s mechanostat theory propose that bone 98
strength is regulated by modeling and remodeling processes which depend on the forces acting on the 99
bones [12]. The mechanical load to bone is increased not only because of physical activity and 100
increased muscle mass but also due to increased FM and particularly obesity [3].
101 102
In addition to the mechanical load, adipose tissue may influence bone metabolism through adipokines, 103
other cytokines, and hormones [13–15]. Adipose tissue may stimulate bone formation by producing 104
estrogens from steroid precursors and by increasing circulating leptin and insulin levels [13–15].
105
However, adipose tissue produces adiponectin and inflammation-related cytokines, such as tumor 106
necrosis factor α (TNF-α) and interleukin 6 (IL-6), which may have deleterious effects on bone [13–
107
15]. Vitamin D is a prohormone converted in the liver to 25-hydroxyvitamin D (25[OH]D) and then 108
in the kidney to 1,25-dihydroxyvitamin D (1,25[OH]²D), the active metabolite which regulates 109
calcium, phosphorus, and bone metabolism [16]. Obesity has been associated with lower serum levels 110
of 25(OH)D [17], that could therefore be one of the links between obesity and BMD.
111 112
More recently, also skeletal muscle and bone have been recognized as endocrine organs [18,19].
113
Skeletal muscle produces myokines, such as myostatin, insulin-like growth factor I (IGF-1), irisin, 114
and IL-6, which may be important mediators in the interaction between skeletal muscle and bone 115
[18,19]. IGF-1 may be one of the factors that mediate the response of bone and skeletal muscle to 116
mechanical loading [19,20]. Osteocytes also secrete IL-6, IGF-1, and other hormone-like factors, 117
5 such as osteocalcin and fibroblast growth factor 23, which have been suggested to play a role in the 118
association between skeletal muscle and bone metabolism [18,19].
119 120
Low BMD in childhood tends to persist until young adulthood [21], and bone mass attained during 121
childhood and adolescence is one of the most important determinants of lifelong skeletal health [22].
122
Pediatric obesity is a growing global health problem [23], and it is therefore important to know how 123
adiposity and associated increase in LM affects BMD among children. There is no consensus on the 124
associations of FM and LM with BMD or the underlying mechanisms. We therefore studied the 125
associations of LM, FM, and associated biomarkers, including adipokines, myokines, inflammation- 126
related biomarkers, growth factors, and 25(OH)D, with BMD assessed by dual-energy x-ray 127
absorptiometry (DXA) in a population sample of children 6-8 years of age.
128
2. Methods 129
2.1 Study design and participants 130
The present analyses are based on the baseline data of the Physical Activity and Nutrition in Children 131
(PANIC) Study, which is an ongoing physical activity and dietary intervention study in a population 132
sample of children 6–8 years of age from the city of Kuopio, Finland (ClinicalTrials.gov registration 133
number NCT01803776). Altogether 736 children from the primary schools of Kuopio were invited 134
to participate in the baseline examinations in 2007—2009. Of the invited children, 512 (70%) 135
participated in the baseline examinations. The participants did not differ in age, sex distribution, or 136
body mass index standard deviation score (BMI-SDS) from all children who started the 1st grade in 137
the city of Kuopio in 2007–2009 based on data from the standard school health examinations. From 138
the present analyses, we excluded children who had chronic diseases or medications that could affect 139
BMD, such as juvenile arthritis demanding long-term treatment with oral corticosteroids. We also 140
excluded 12 children who had entered puberty to avoid associated confounding. Complete data on 141
the main variables used in the present analyses were available for 472 children (227 girls, 245 boys).
142
The study was conducted according to the ethical guidelines laid down in the Declaration of Helsinki.
143
The study protocol was approved by the Research Ethics Committee of the Hospital District of 144
Northern Savo. Both children and their parents gave their written informed consent.
145
6 2.2 Assessment of bone mineral density and body composition
146
LM, FM, body fat percentage (BF %), and BMD of the whole body excluding the head were assessed 147
using the Lunar Prodigy Advance® DXA device (GE Medical Systems, Madison, WI, USA) and the 148
Encore® software, Version 10.51.006 (GE Company, Madison, WI, USA), according to the 149
manufacturer’s instructions using standardized protocols. The same DXA device and software were 150
used in all measurements. Body weight was measured twice the children having fasted for 12 hours, 151
emptied the bladder, and standing in light underwear by the InBody® 720 bioelectrical impedance 152
device (Biospace, Seoul, Korea) to accuracy of 0.1 kg. The mean of these two values was used in the 153
analyses. Body height was measured three times the children standing in the Frankfurt plane without 154
shoes using a wall-mounted stadiometer to accuracy of 0.1 cm. The mean of the nearest two values 155
was used in the analyses. BMI-SDS was calculated using national reference values [24]. Waist 156
circumference was measured three times after expiration at mid-distance between the bottom of the 157
rib cage and the top of the iliac crest with an unstretchable measuring tape to accuracy of 0.1 cm. The 158
mean of the nearest two values was used in the analyses. Intraclass correlation coefficients for body 159
weight and height and waist circumference were >0.99.
160
2.3 Biochemical analyses 161
Venous blood samples were taken the children having fasted for 12 hours. Blood was immediately 162
centrifuged and stored at a temperature of -75ºC until biochemical analyses, except for glucose that 163
was measured from non-frozen plasma samples. Serum 25(OH)D concentration was analysed by a 164
chemiluminescence immunoassay called the LIAISON® 25 OH Vitamin D TOTAL Assay (DiaSorin 165
Inc., Stillwater, USA) as described earlier [25,26]. Serum dehydroepiandrosterone sulphate (DHEAS) 166
concentration was used as a marker of biochemical adrenarche and was determined using an enzyme 167
linked immunosorbent assay (ELISA) kit (Alpha Diagnostic International, San Antonio, Texas, USA) 168
[27]. Serum IGF-1 concentration was analysed using an ELISA kit (Mediagnost, Reutlingen, 169
Germany). Plasma glucose concentration was measured using the hexokinase method (Roche 170
Diagnostics GmbH, Mannheim, Germany). Serum insulin concentration was measured by the 171
electrochemiluminescence immunoassay with the sandwich principle (Roche Diagnostics GmbH, 172
Mannheim, Germany). We calculated the Homeostatic Model Assessment for Insulin Resistance 173
(HOMA-IR) using the formula fasting serum insulin x fasting plasma glucose/22. Serum high- 174
molecular-weight adiponectin concentration was analysed using an ELISA kit after a specific 175
proteolytic digestion of other multimeric adiponectin forms (Millipore, Billerica, MA, USA). Plasma 176
leptin concentration was measured by a competitive radioimmunoassay (Multigamma 1261-001, 177
7 PerkinElmer Wallac Oy, Turku, Finland) and plasma soluble leptin receptor concentration using an 178
ELISA kit (Multicalc evaluation programme PerkinElmer Wallac Oy, Turku, Finland). We calculated 179
the free leptin index by dividing leptin with soluble leptin receptor and multiplying by 100 [28].
180
Commercially available ELISA kits were employed for the measurement of plasma irisin (Phoenix 181
Pharmaceuticals, Burlingame, California, USA), IL-6, and TNF-α concentrations (Sanquin Reagents, 182
Amsterdam, The Netherlands). Plasma high-sensitivity C-reactive protein (hsCRP) was measured 183
using an enhanced immunoturbidimetric assay with the CRP (Latex) High Sensitive Assay reagent 184
(Roche Diagnostics GmbH, Mannheim, Germany) and the limit of quantitation of 0.3 mg/l.
185
2.4 Assessments of general health, puberty, and adrenarche 186
The parents filled out a questionnaire that included items on the children’s chronic diseases and 187
allergies diagnosed by a physician as well as detailed information on the children’s use of 188
medications. A research physician assessed pubertal status during a medical examination. Central 189
puberty was defined as breast development at Tanner stage ≥2 for girls and testicular volume ≥4 mL 190
assessed using an orchidometer for boys. Premature adrenarche was defined as serum DHEAS ≥ 1 191
µmol/l (≥ 37 µg/dl) [29] and at least one clinical sign of androgen action. Birth weight was obtained 192
from Kuopio University Hospital record, and birth weight -SDS was calculated according to Finnish 193
growth reference data [30].
194
2.5 Statistical methods 195
We performed statistical analyses using the IBM SPSS Statistics® software, Version 21 (IBM Corp., 196
Armonk, NY, USA). The normality of distributions of the variables was verified visually and by the 197
Kolmogorov-Smirnov test. The t-test for independent samples and the Mann–Whitney’s U-test were 198
used to examine differences in the basic characteristics between sexes. Linear regression analysis was 199
used to investigate the determinants of BMD, and the normality of residuals for regression models 200
was assessed using histograms. Model 1 included each determinant of BMD separately, adjusted for 201
age and sex. Model 2 was additionally adjusted for body height. Model 3 included all variables in 202
Model 2 and LM, and Model 4 included all variables in Model 2 and FM. Corresponding linear 203
regression analyses were also performed for girls and boys separately. FM had a strong positive 204
correlation with leptin in girls (r=0.789, p<0.001), boys (r=0.850, p<0.001), and girls and boys 205
combined (r=0.810, p<0.001). We therefore tested whether FM modified the association between 206
leptin and BMD by analyzing this association in the sex-specific thirds of FM using linear regression 207
8 analysis adjusted for age, sex, and body height. In all analyses, associations with a p-value of <0.05 208
were considered statistically significant.
209
3. Results 210
3.1 Characteristics of children 211
The boys were heavier and taller and had higher waist circumference and LM and lower BF% and 212
FM than the girls, but there was no difference in BMI-SDS between the genders (Table 1). The girls 213
had higher IGF-1, insulin, leptin, and free leptin index and lower leptin receptor and IL-6 than the 214
boys. Of the children, 38 (8.1%) had asthma, 128 (27.1%) any allergic symptom (rhinitis, 215
conjunctivitis, atopy, food or medicine allergy), 21 (4.4%) an attention deficit hyperactivity disorder 216
(ADHD/ADD) or another mild neurocognitive disorder or developmental delay, 8 (1.7%) a mild 217
congenital dysmorphism, and 10 (2.1%) any other chronic disease. There was no difference in BMD 218
between children with these diseases and those without them.
219
3.2. Determinants of bone mineral density in all children 220
Body height (β=0.572, p<0.001) and weight (β=0.709, p<0.001) were positively associated with 221
BMD adjusted for age and sex. LM was also a strong positive correlate for BMD adjusted for age and 222
sex (Table 2, Model 1). This association remained similar after additional adjustment for body height 223
(Model 2) but weakened slightly after further adjustment for FM (Model 4). Moreover, FM had a 224
strong positive association with BMD adjusted for age and sex (Table 2, Model 1). This association 225
weakened after additional adjustment for body height (Model 2) but remained similar when further 226
adjusted for LM (Model 3). Birth weight was positively associated with BMD adjusted for age and 227
sex (Table 2, Model 1), but this association disappeared after additional adjustments (Models 2-4).
228 229
Serum 25(OH)D was positively associated with BMD adjusted for age and sex (Table 2, Model 1).
230
This association remained almost similar after additional adjustment for body height and FM (Models 231
2 and 4) but was no longer statistically significant when adjusted for LM (Model 3). DHEAS was 232
positively associated with BMD adjusted for age and sex (Table 2, Model 1). This association 233
weakened when additionally adjusted for body height (Model 2) but was no longer statistically 234
significant after adjustment for LM or FM (Models 3-4). IGF-1 was a positive correlate for BMD 235
adjusted for age and sex (Table 2, Model 1) but not after further adjustments (Models 2-4). Insulin 236
and HOMA-IR were positively associated with BMD adjusted for age and sex (Table 2, Model 1).
237
9 These associations weakened after additional adjustment for body height (Model 2). The association 238
of insulin weakened and that of HOMA-IR was no longer statistically significant after further 239
adjustment for LM (Model 3). The associations of insulin and HOMA-IR with BMD disappeared 240
when adjusted for FM (Model 4).
241 242
Adiponectin was a negative correlate for BMD adjusted for age and sex (Table 2, Model 1) but not 243
after further adjustments (Models 2-4). Leptin was positively associated with BMD adjusted for age 244
and sex (Table 2, Model 1). This association weakened after additional adjustment for body height 245
and LM (Models 2-3) and was no longer statistically significant after adjustment for FM (Model 4).
246
There was a positive association between leptin and BMD in the highest sex-specific third of FM 247
(β=0.274, p<0.001) but a non-significant inverse association in the middle third (β=-0.144, p=0.058) 248
and the lowest third (β=-0.112, p=0.118) adjusted for age and body height. Lower leptin receptor and 249
higher free leptin index were associated with higher BMD adjusted for age and sex (Table 2, Model 250
1). These associations weakened after additional adjustment for body height and when further 251
adjusted for LM (Models 2-3) and were no longer statistically significant after adjustment for FM 252
(Model 4). Irisin was positively associated with BMD adjusted for age and sex (Table 2, Model 1).
253
This association weakened slightly when additionally adjusted for body height (Model 2) and 254
remained similar after further adjustment for LM or FM (Models 3-4).
255 256
IL-6 and TNF-α were not associated with BMD (Table 2, Models 1-4). Higher hs-CRP was associated 257
with higher BMD adjusted for age and sex (Table 2, Model 1), after additional adjustment for body 258
height (Model 2), and also when further adjusted for LM (Model 3). However, this association 259
disappeared after adjustment for FM (Model 4).
260
3.2.2 Determinants of bone mineral density in girls 261
In girls, body height (β=0.615, p<0.001) and weight (β=0.727, p<0.001) were positively associated 262
with BMD adjusted for age. LM had a strong positive association with BMD adjusted for age, body 263
height, and FM (Table 3, Models 1, 2, and 4). FM was also a strong positive correlate for BMD 264
adjusted for age, body height, and LM (Table 3, Models 1-3). Birth weight SDS, 25(OH)D, DHEAS, 265
IGF-1, and irisin were positively associated with BMD when adjusted for age (Table 3, Model 1) but 266
not after further adjustments (Models 2-4). Insulin and HOMA-IR were positive correlates for BMD 267
adjusted for age, body height, and LM (Table 3, Models 1-3) but not when adjusted for FM (Model 268
4). Leptin and free leptin index were positively and leptin receptor was negatively associated with 269
BMD adjusted for age, body height, and LM (Table 3, Models 1-3) but not adjusted for FM (Model 270
10 4). There was a positive association between leptin and BMD in the highest third of FM (β=0.346, 271
p<0.001) but a non-significant inverse association in the middle third (β=-0.169, p=0.126) and the 272
lowest third (β=-0.122, p=0.261) adjusted for age and body height.
273
3.2.3 Determinants of bone mineral density in boys 274
In boys, body height (β=0.520, p<0.001) and weight (β=0.686, p<0.001) were positively associated 275
with BMD adjusted for age. LM had a strong positive association with BMD adjusted for age, body 276
height, and FM (Table 4, Models 1, 2, and 4). FM was also a strong positive correlate for BMD 277
adjusted for age, body height, and LM (Table 4, Models 1-3). Serum 25(OH)D was positively 278
associated with BMD adjusted for age, body height, and FM (Table 4, Models 2 and 4) but not 279
adjusted for LM (Model 4). Birth weight SDS, DHEAS, insulin, HOMA-IR and hs-CRP were 280
positively associated with BMD adjusted for age (Table 4, Model 1) but not after further adjustments 281
(Models 2-4). IGF-1 was negatively associated with BMD only when adjusted for age, body height, 282
and FM (Table 4, Model 4). Leptin and free leptin index were positively and leptin receptor was 283
negatively associated with BMD adjusted for age, body height, and LM (Table 4, Models 1-3), but 284
the associations of free leptin index and leptin receptor were no longer statistically significant and 285
that of leptin became negative when adjusted for LM (Model 4). There was a non-significant positive 286
association between leptin and BMD in the highest third of FM (β=0.199, p=0.061), a non-significant 287
inverse association in the middle third (β=-0.135, p=0.203) and no association in the lowest third (β=- 288
0.024, p=0.821).
289
4. Discussion 290
Our study is one of the few studies on the associations of LM, FM, and various biomarkers secreted 291
by adipose tissue, skeletal muscle, or bone with BMD in a population sample of prepubertal children.
292
LM but also FM were strong and independent positive determinants of BMD in all children, girls, 293
and boys. Plasma irisin was also an independent positive correlate for BMD in all children but not in 294
girls and boys separately. The associations of other biomarkers were explained by body height, LM, 295
or FM. In boys, the positive association between leptin and BMD became negative and the negative 296
association between IGF-1 and BMD strengthened after controlling for FM.
297 298
In line with previous studies among children and adolescents [4,5,7], LM was a strong positive 299
correlate for BMD in the current study. The positive association between LM and BMD may be 300
11 explained by increased mechanical load to bone caused by increased LM and the loading effect of 301
weight-bearing exercise on bone mass and metabolism [11].
302 303
A recently identified myokine irisin is produced by skeletal muscle after exercise and may increase 304
energy expenditure [31]. Irisin has been found to increase bone mass in mice [32], but evidence on 305
the association between serum irisin and BMD in humans is limited. Irisin has been positively 306
associated with bone mass and strength in young athletes and negatively related to vertebral fragility 307
fractures in postmenopausal women [31,33]. To the best of our knowledge, the association between 308
irisin and BMD has not been studied earlier in children. We found that higher serum irisin levels were 309
associated with higher BMD even after controlling for LM or FM. The weak positive association 310
between irisin and BMD was slightly stronger in girls than in boys, but statistical power was limited 311
in these sex-specific analyses.
312 313
Of other biomarkers previously related to skeletal muscle and bone metabolism, insulin had a weak 314
positive association with BMD even after controlling for LM. However, the association between 315
insulin and BMD was explained by FM. IGF-1 was positively associated with BMD in all children 316
and in girls but not after controlling for body size and composition. Moreover, there was a weak 317
negative association between IGF-1 and BMD in boys when controlled for FM. Previous studies in 318
children and adolescents have reported an independent positive association between IGF-1 and bone 319
growth [20] and a muscle-dependent positive association between IGF-1 and BMD [20,34]. However, 320
insulin resistance has suppressed the muscle-dependent relationship between IGF-1 and BMC and 321
cortical bone measurements in children 9-13 years of age [34,35]. One reason for the inconsistency 322
between our results and the findings of earlier studies could be that our participants were prepubertal 323
and slightly younger than those of the previous studies. It is also possible that the weak negative 324
association between IGF-1 and BMD in boys after controlling for FM in our study is partly explained 325
by the positive relationships among adiposity, insulin resistance, and IGF-1.
326 327
FM has been positively associated with BMD in some previous studies among mainly prepubertal 328
children [6,36]. Obesity has also been associated with increased bone mass independent of LM in a 329
study among children and adolescents [37]. Moreover, adiposity was associated with increased bone 330
mass in another study in adolescents, but this association was explained by LM [7]. One explanation 331
for the positive association between FM and BMD among children and adolescents could be the 332
increased mechanical load to the bone due to adiposity [3]. Another reason could be that adipose 333
tissue stimulates bone growth [36]. However, one study reported a decreased volumetric BMD in 334
12 obese prepubertal children despite increased bone size [38]. Another study showed an inverse 335
association between BF% and BMD in adolescents [5]. In a Finnish study among prepubertal and 336
pubertal children, those with decreased body fat content and those with increased fat content had 337
decreased BMD independent of LM [39]. In the current study, FM was positively associated with 338
BMD independent of LM, even though LM was a stronger correlate for BMD than FM. This 339
observation is consistent with the results of a previous study among children [6]. Studies that have 340
shown an association between excess fat mass and decreased BMD have been conducted in older and 341
more overweight children and adolescents [5,39] than the participants of our study. Only 14% of the 342
girls and 10% of the boys in our population sample of prepubertal children 6-8 years of age were 343
overweight or obese [40]. Therefore, we cannot draw a conclusion on the association between obesity 344
and BMD based on our findings. It is possible that the detrimental effect of excess fat mass appears 345
in later childhood or in adolescence during or after puberty along with changes in body composition 346
[1]. In our study, the association between LM and BMD was stronger in boys than in girls. One reason 347
for this finding could be that boys have more skeletal muscle and girls have more adipose tissue 348
already in prepubertal stage [1], that is consistent with our observation.
349 350
Leptin is an adipocyte-secreted hormone that decreases appetite and increases energy expenditure 351
[14] but may also influence bone modeling through central and peripheral mechanisms [14,15].
352
Leptin has been suggested to inhibit bone formation indirectly through the sympathetic nervous 353
system [14,15]. In contrast, leptin directly enhances bone formation and inhibits bone resorption 354
peripherally, even though the mechanisms are rather complex and not yet well defined [14,15]. These 355
local effects of leptin on bone have been suggested to be dominant, and higher circulating leptin levels 356
may therefore be related to a stronger skeleton [15]. Leptin may also regulate the hypothalamic- 357
pituitary-peripheral endocrine axes, including thyroid, gonadal, cortisol, and growth hormone axes, 358
which are possible additional indirect ways by which leptin affects bone [41]. Soluble leptin receptor 359
is the major protein binding leptin in blood, and leptin receptor levels seem to vary independent of 360
serum leptin levels during childhood [28]. Functional differences between free and bound leptin are 361
not clear, but some studies have suggested that free leptin index better reflects the physiological 362
actions of leptin [28]. A meta-analysis concluded that circulating leptin levels were positively 363
associated with BMD [42], but most of the 46 studies included in the analysis were performed in 364
adults. The association between leptin and total body BMD was also positive in five studies among 365
girls [42]. Interestingly, the relationship between leptin and BMD adjusted for body mass was 366
negative in the only small study among boys [43]. Furthermore, body fat content was not taken into 367
account in the meta-analysis [42]. In a previous study, free leptin index was associated with bone 368
13 turnover markers [13], which may be one mechanism for the inverse association between leptin and 369
BMD. We found that leptin receptor level was negatively and leptin and free leptin index were 370
positively associated with BMD independent of LM, but these associations were explained by FM.
371
Moreover, the association between leptin and BMD became negative in boys after controlling for 372
FM. Leptin was positively associated with BMD in the highest sex-specific third of FM but had a 373
weak negative association in the middle and lowest thirds. These findings suggest that FM strongly 374
modifies the association between leptin and BMD.
375 376
Adiponectin is an adipokine that has been inversely related to FM in children [44], and this inverse 377
association has been found to strengthen in puberty [45]. Adiponectin regulates energy homeostasis, 378
glucose and lipid metabolism, and inflammatory pathways [15]. Increased adiponectin has been 379
associated with reduced bone mass in children [44]. This may be explained by the decreased 380
circulating levels of insulin and IGF-1 due to increased adiponectin levels [15]. In the current study 381
among prepubertal children, we found a weak negative association between adiponectin and BMD, 382
but it was largely explained by LM and FM. It is possible that the negative association between 383
adiponectin and BMD might be stronger after puberty.
384 385
Excess adiposity is associated with insulin resistance and hyperinsulinemia in youth [46]. Insulin has 386
been suggested to be anabolic for bone formation, and higher serum insulin levels have been 387
associated with higher BMD in adults [15]. However, the associations of insulin resistance with BMC 388
and BMD remain controversial in children and adolescents [47–49]. In a study among prepubertal 389
overweight children, BMC was lower in children with prediabetes than in children without it [47]. In 390
overweight adolescents, increased HOMA-IR was associated with decreased BMD [48]. In another 391
study among adolescents, insulin was positively associated with BMD, but the association was 392
inverse after controlling for FM [49]. In line with these results, we found that higher fasting insulin 393
and HOMA-IR were associated with higher BMD, but the associations became weak negative in boys 394
and disappeared in girls after controlling for FM. These findings suggest that the association between 395
insulin resistance and BMD is largely dependent on adiposity that should be taken into account when 396
interpreting the results.
397 398
IL-6 has a double-edged role in bone metabolism as it may stimulate both osteocyte differentiation 399
and osteoclastic bone resorption [19]. IL-6 but also TNF-α are inflammation-related cytokines 400
secreted by adipose tissue, and they may enhance bone resorption [14]. We found no association 401
between IL-6 or TNF-α and BMD in children. One explanation for this may be that the prevalence of 402
14 overweight was low in our general population of children, and thus the inflammatory-related effects 403
of these cytokines may have been modest. Higher hs-CRP has been associated with lower BMD in 404
adolescent girls [50] and in overweight children with prediabetes but not in overweight children 405
without it [47]. Inconsistent with these findings, we observed a weak positive association between 406
hs-CRP and BMD in children. The reason for this inconsistency probably is the low proportion of 407
overweight and obese children in our population sample [40]. Moreover, the observed positive 408
association between hs-CRP and BMD was explained by FM. This is an expected result as adiposity 409
is known to be related to systemic low-grade inflammation [51].
410 411
The definition of vitamin D deficiency based on serum 25(OH)D concentration varies between 25 412
and 50 nmol/l and the lower limit for optimal serum 25(OH)D concentration has been suggested to 413
be as high as 75 nmol/l [3,16,52–57]. No consensus exists on the optimal serum level of 25(OH)D.
414
As vitamin D is essential for bone metabolism [16], the positive association of 25(OH)D with BMD 415
in the current study was expected, and this is in line with the results of previous studies [4]. However, 416
the association between 25(OH)D and BMD was weak especially in girls, but this is probably 417
explained by the low proportion of children having 25(OH)D concentrations below 50 nmol/l [25], 418
which has been considered as a limit of deficiency based on bone outcomes [53]. The association 419
between 25(OH)D and BMD was stronger in boys, and it was partly explained by LM. One 420
explanation for this finding may be that physically active children, particularly boys, have increased 421
LM and spend more time outdoors and are therefore exposed to sunlight that increases serum 422
25(OH)D concentrations.
423 424
DHEAS is an androgen precursor produced mainly by the adrenal cortex and whose circulating levels 425
are increased during adrenarche [27]. Both obesity and premature adrenarche are associated with 426
advanced bone age [58,59]. However, there are little and inconsistent data on the association between 427
DHEAS and BMD in children [58,60]. In the current study among prepubertal children, higher 428
DHEAS was associated with higher BMD. However, the positive association weakened after 429
controlling for body height, LM, and FM, suggesting that DHEAS does not have an independent 430
effect on BMD in prepubertal children.
431 432
Some diseases, conditions and medications, such as juvenile arthritis, renal insufficiency, 433
inflammatory conditions, disabilities, immobility, oral corticosteroid use, or certain antiepileptic 434
drugs, may decrease BMD [61]. We therefore excluded children who had such diseases, conditions, 435
or medications to avoid associated confounding. The use of inhaled corticosteroids has been 436
15 associated with decreased BMD in some studies [62]. However, a recent review and meta-analysis 437
concluded that the use of inhaled corticosteroids was not associated with decreased lumbar BMD or 438
increased risk of fractures [63]. In our study, about 8% of the children had asthma, a few of them used 439
regular inhaled corticosteroids, and they had similar BMD to children without asthma. We therefore 440
included children with asthma in the current study population.
441 442
Body weight and BMI have been directly associated with BMD in children and adolescents [3,6], but 443
neither of them is a specific measure of LM or FM. We therefore investigated the associations of LM 444
and FM measured by DXA with BMD among children. DXA is also the most widely used method to 445
evaluate BMD and it has been reported to be well reproducible also in children [64–66]. The 446
assessment and interpretation of BMD measurements are not simple in growing children because of 447
both methodological aspects and differences in maturation and growth. In children, The International 448
Society of Clinical Densitometry (ISCD) recommends measuring BMD and BMC from total body 449
excluding the head and the posterior-anterior spine [66]. Areal BMD measurements may 450
underestimate the BMD of short children and overestimate the BMD of tall children. Therefore, ISCD 451
recommends adjusting BMD of total body excluding head and spinal BMD using height z-score. We 452
used DXA of the whole body, excluding the head, which is one of the methods recommended to be 453
used for measuring BMD among children by the ISCD [66]. Moreover, we adjusted the data first for 454
age and sex and then additionally for body height, all components of height z-score. However, we did 455
not measure volumetric BMD but areal BMD and did not use computed tomography to measure the 456
more detailed quality of the bone.
457 458
The results of different studies depend not only on the methods used but also on the age and 459
maturation of the participants and the prevalence of overweight in the study population, because each 460
of them affects BMD. We investigated a general population of prepubertal children 6-8 years of age 461
with a low prevalence of overweight, whereas many other studies have mainly included overweight 462
or obese children and adolescents with advanced puberty [5,7,37,39,47]. It is therefore difficult to 463
compare the findings of our study with those of many other studies.
464
5. Conclusions 465
Our study showed that LM is the strongest positive determinant of BMD, but also FM is positively 466
and independently associated with BMD in a population sample of mainly normal-weight prepubertal 467
Finnish children. Of biomarkers related to body composition, irisin had a positive association with 468
BMD independent of LM and FM. To the best of our knowledge, this is the first study to examine the 469
16 association between irisin and BMD in children, and this finding needs to be confirmed in other 470
populations. As expected, 25(OH)D was a positive correlate for BMD, but the association was weak 471
probably due to the relatively low prevalence of vitamin D deficiency in our study population and 472
was partly explained by body composition. In boys, the positive association of leptin with BMD 473
became negative after controlling for FM. This finding suggests that FM strongly modifies the 474
association between leptin and BMD and that adiposity should be taken into account when 475
interpreting the associations of leptin with bone structure and metabolism.
476
6. Acknowledgements 477
The authors are grateful to all the children and their parents for participating in the PANIC study. The 478
authors are also indebted to the members of the PANIC research team for their skillful contribution 479
in performing the study. The authors are grateful to Ayhan Korkmaz for performing irisin 480
measurements, Leila Antikainen for performing DHEAS and IGF-1 measurements, Tuomas Onnukka 481
for performing leptin measurements, and Kaija Kettunen for performing leptin receptor and 482
adiponectin measurements. We also thank Tarja Kokkola for the help with methodological issues on 483
laboratory measurements.
484
7. Funding sources 485
This work was financially supported by grants from Ministry of Social Affairs and Health of Finland, 486
Ministry of Education and Culture of Finland, Finnish Innovation Fund Sitra, Social 487
Insurance Institution of Finland, Finnish Cultural Foundation, Juho Vainio Foundation, Foundation 488
for Paediatric Research, Doctoral Programs in Public Health, Paavo Nurmi Foundation, 489
Paulo Foundation, Diabetes Research Foundation, Yrjö Jahnsson Foundation, Finnish Foundation for 490
Cardiovascular Research, Research Committee of the Kuopio University Hospital Catchment Area 491
(State Research Funding), Kuopio University Hospital (previous state research funding (EVO), 492
funding number 5031343), and the city of Kuopio.
493
8. Conflict of interest 494
The authors declare there are no conflicts of interest.
495
17 9. References
496
[1] L.A. Loomba-Albrecht, D.M. Styne, Effect of puberty on body composition, Curr. Opin.
497
Endocrinol. Diabetes. Obes. 16 (2009) 10–15. doi:10.1097/MED.0b013e328320d54c.
498
[2] A.B. Sopher, I. Fennoy, S.E. Oberfield, An update on childhood bone health: mineral accrual, 499
assessment and treatment, Curr. Opin. Endocrinol. Diabetes. Obes. 22 (2015) 35–40.
500
doi:10.1097/MED.0000000000000124.
501
[3] N.H. Golden, S.A. Abrams, Optimizing bone health in children and adolescents, Pediatrics.
502
134 (2014) e1229–e1243. doi:10.1542/peds.2014-2173.
503
[4] M. Pekkinen, H. Viljakainen, E. Saarnio, C. Lamberg-Allardt, O. Mäkitie, Vitamin D is a 504
major determinant of bone mineral density at school age, PLoS One. 7 (2012) e40090.
505
doi:10.1371/journal.pone.0040090.
506
[5] L.N. Mosca, T.B.L. Goldberg, V.N. da Silva, C.C. da Silva, C.S. Kurokawa, A.C. Bisi Rizzo, 507
J.E. Corrente, Excess body fat negatively affects bone mass in adolescents, Nutrition. 30 508
(2014) 847–852. doi:10.1016/j.nut.2013.12.003.
509
[6] M. Heidemann, R. Holst, A.J. Schou, H. Klakk, S. Husby, N. Wedderkopp, C. Molgaard, The 510
influence of anthropometry and body composition on children’s bone health: the childhood 511
health, activity and motor performance school (the CHAMPS) study, Denmark, Calcif. Tissue 512
Int. 96 (2015) 97–104. doi:10.1007/s00223-014-9941-9.
513
[7] L. Gracia-Marco, F.B. Ortega, D. Jiménez-Pavón, G. Rodríguez, M.J. Castillo, G. Vicente- 514
Rodríguez, L.A. Moreno, Adiposity and bone health in Spanish adolescents. The HELENA 515
study., Osteoporos. Int. 23 (2012) 937–47. doi:10.1007/s00198-011-1649-3.
516
[8] P. Dimitri, N. Bishop, J.S. Walsh, R. Eastell, Obesity is a risk factor for fracture in children 517
but is protective against fracture in adults: A paradox, Bone. 50 (2012) 457–466.
518
doi:10.1016/j.bone.2011.05.011.
519
[9] H.A. Weiler, L. Janzen, K. Green, J. Grabowski, M.M. Seshia, K.C. Yuen, Percent body fat 520
and bone mass in healthy Canadian females 10 to 19 years of age, Bone. 27 (2000) 203–207.
521
doi:10.1016/S8756-3282(00)00314-8.
522
[10] A. Goulding, A.M. Grant, S.M. Williams, Bone and body composition of children and 523
adolescents with repeated forearm fractures, J. Bone Miner. Res. 20 (2005) 2090–2096.
524
doi:10.1359/JBMR.050820.
525
[11] K. Hind, M. Burrows, Weight-bearing exercise and bone mineral accrual in children and 526
adolescents: A review of controlled trials, Bone. 40 (2007) 14–27.
527
doi:10.1016/j.bone.2006.07.006.
528
18 [12] H.M. Frost, Bone “mass” and the “mechanostat”: a proposal., Anat. Rec. 219 (1987) 1–9.
529
doi:10.1002/ar.1092190104.
530
[13] P. Dimitri, J.K. Wales, N. Bishop, Adipokines, bone-derived factors and bone turnover in 531
obese children; evidence for altered fat-bone signalling resulting in reduced bone mass, Bone.
532
48 (2011) 189–196. doi:10.1016/j.bone.2010.09.034.
533
[14] M. Kawai, F.J.A. de Paula, C.J. Rosen, New insights into osteoporosis: The bone-fat 534
connection, J. Intern. Med. 272 (2012) 317–329. doi:10.1111/j.1365-2796.2012.02564.x.
535
[15] I.R. Reid, Fat and bone., Arch. Biochem. Biophys. 503 (2010) 20–7.
536
doi:10.1016/j.abb.2010.06.027.
537
[16] M.F. Holick, Vitamin D Deficiency, N. Engl. J. Med. 357 (2007) 266–281.
538
doi:10.1056/NEJMra070553.
539
[17] C.E. Moore, Y. Liu, Low serum 25-hydroxyvitamin D concentrations are associated with total 540
adiposity of children in the United States: National Health and Examination Survey 2005 to 541
2006., Nutr. Res. 36 (2016) 72–9. doi:10.1016/j.nutres.2015.11.003.
542
[18] M. Brotto, L. Bonewald, Bone and muscle: Interactions beyond mechanical., Bone. 80 (2015) 543
109–14. doi:10.1016/j.bone.2015.02.010.
544
[19] A.D. Bakker, R.T. Jaspers, IL-6 and IGF-1 Signaling Within and Between Muscle and Bone:
545
How Important is the mTOR Pathway for Bone Metabolism?, Curr. Osteoporos. Rep. 13 546
(2015) 131–9. doi:10.1007/s11914-015-0264-1.
547
[20] L. Xu, Q. Wang, Q. Wang, A. Lyytikäinen, T. Mikkola, E. Völgyi, S. Cheng, P. Wiklund, E.
548
Munukka, P. Nicholson, M. Alén, S. Cheng, Concerted actions of insulin-like growth factor 1, 549
testosterone, and estradiol on peripubertal bone growth: a 7-year longitudinal study., J. Bone 550
Miner. Res. 26 (2011) 2204–11. doi:10.1002/jbmr.422.
551
[21] T.A.L. Wren, H.J. Kalkwarf, B.S. Zemel, J.M. Lappe, S. Oberfield, J.A. Shepherd, K.K.
552
Winer, V. Gilsanz, Longitudinal tracking of dual-energy X-ray absorptiometry bone measures 553
over 6 years in children and adolescents: Persistence of low bone mass to maturity, J. Pediatr.
554
164 (2014) 1280–1285.e2. doi:10.1016/j.jpeds.2013.12.040.
555
[22] NIH Consensus Development Panel on Osteoporosis Prevention, Diagnosis, and Therapy.
556
Osteoporosis Prevention, Diagnosis, and Therapy, JAMA J. Am. Med. Assoc. 285 (2001) 785–
557
795. doi:10.1001/jama.285.6.785.
558
[23] M. Ng, T. Fleming, M. Robinson, B. Thomson, N. Graetz, C. Margono, E.C. Mullany, S.
559
Biryukov, C. Abbafati, S.F. Abera, J.P. Abraham, N.M.E. Abu-Rmeileh, T. Achoki, F.S.
560
AlBuhairan, Z.A. Alemu, R. Alfonso, M.K. Ali, R. Ali, N.A. Guzman, W. Ammar, P. Anwari, 561
A. Banerjee, S. Barquera, S. Basu, D.A. Bennett, Z. Bhutta, J. Blore, N. Cabral, I.C. Nonato, 562
19 J.-C. Chang, R. Chowdhury, K.J. Courville, M.H. Criqui, D.K. Cundiff, K.C. Dabhadkar, L.
563
Dandona, A. Davis, A. Dayama, S.D. Dharmaratne, E.L. Ding, A.M. Durrani, A. Esteghamati, 564
F. Farzadfar, D.F.J. Fay, V.L. Feigin, A. Flaxman, M.H. Forouzanfar, A. Goto, M.A. Green, 565
R. Gupta, N. Hafezi-Nejad, G.J. Hankey, H.C. Harewood, R. Havmoeller, S. Hay, L.
566
Hernandez, A. Husseini, B.T. Idrisov, N. Ikeda, F. Islami, E. Jahangir, S.K. Jassal, S.H. Jee, 567
M. Jeffreys, J.B. Jonas, E.K. Kabagambe, S.E.A.H. Khalifa, A.P. Kengne, Y.S. Khader, Y.-H.
568
Khang, D. Kim, R.W. Kimokoti, J.M. Kinge, Y. Kokubo, S. Kosen, G. Kwan, T. Lai, M.
569
Leinsalu, Y. Li, X. Liang, S. Liu, G. Logroscino, P.A. Lotufo, Y. Lu, J. Ma, N.K. Mainoo, 570
G.A. Mensah, T.R. Merriman, A.H. Mokdad, J. Moschandreas, M. Naghavi, A. Naheed, D.
571
Nand, K.M.V. Narayan, E.L. Nelson, M.L. Neuhouser, M.I. Nisar, T. Ohkubo, S.O. Oti, A.
572
Pedroza, D. Prabhakaran, N. Roy, U. Sampson, H. Seo, S.G. Sepanlou, K. Shibuya, R. Shiri, 573
I. Shiue, G.M. Singh, J.A. Singh, V. Skirbekk, N.J.C. Stapelberg, L. Sturua, B.L. Sykes, M.
574
Tobias, B.X. Tran, L. Trasande, H. Toyoshima, S. van de Vijver, T.J. Vasankari, J.L. Veerman, 575
G. Velasquez-Melendez, V.V. Vlassov, S.E. Vollset, T. Vos, C. Wang, X. Wang, E.
576
Weiderpass, A. Werdecker, J.L. Wright, Y.C. Yang, H. Yatsuya, J. Yoon, S.-J. Yoon, Y. Zhao, 577
M. Zhou, S. Zhu, A.D. Lopez, C.J.L. Murray, E. Gakidou, Global, regional, and national 578
prevalence of overweight and obesity in children and adults during 1980–2013: a systematic 579
analysis for the Global Burden of Disease Study 2013, Lancet. 384 (2014) 766–781.
580
doi:10.1016/S0140-6736(14)60460-8.
581
[24] A. Saari, U. Sankilampi, M.-L. Hannila, V. Kiviniemi, K. Kesseli, L. Dunkel, New Finnish 582
growth references for children and adolescents aged 0 to 20 years: Length/height-for-age, 583
weight-for-length/height, and body mass index-for-age., Ann. Med. 43 (2011) 235–248.
584
doi:10.3109/07853890.2010.515603.
585
[25] S. Soininen, A.-M. Eloranta, V. Lindi, T. Venäläinen, N. Zaproudina, A. Mahonen, T.A.
586
Lakka, Determinants of serum 25-hydroxyvitamin D concentration in Finnish children: the 587
Physical Activity and Nutrition in Children (PANIC) study, Br. J. Nutr. 25 (2016) 1–12.
588
doi:10.1017/S0007114515005292.
589
[26] S. Soininen, A.-M. Eloranta, V. Lindi, T.A. Lakka, Response: food fortification as a means to 590
increase vitamin D intake., Br. J. Nutr. 116 (2016) 1–2. doi:10.1017/S0007114516002890.
591
[27] A. Mäntyselkä, J. Jääskeläinen, V. Lindi, A. Viitasalo, T. Tompuri, R. Voutilainen, T.A.
592
Lakka, The presentation of adrenarche is sexually dimorphic and modified by body adiposity., 593
J. Clin. Endocrinol. Metab. 99 (2014) 3889–94. doi:10.1210/jc.2014-2049.
594
[28] J. Kratzsch, A. Lammert, A. Bottner, B. Seidel, G. Mueller, J. Thiery, J. Hebebrand, W. Kiess, 595
Circulating soluble leptin receptor and free leptin index during childhood, puberty, and 596
20 adolescence., J. Clin. Endocrinol. Metab. 87 (2002) 4587–94. doi:10.1210/jc.2002-020001.
597
[29] P. Utriainen, R. Voutilainen, J. Jaaskelainen, Continuum of phenotypes and sympathoadrenal 598
function in premature adrenarche, Eur. J. Endocrinol. 160 (2009) 657–665. doi:10.1530/EJE- 599
08-0367.
600
[30] U. Sankilampi, M.-L. Hannila, A. Saari, M. Gissler, L. Dunkel, New population-based 601
references for birth weight, length, and head circumference in singletons and twins from 23 to 602
43 gestation weeks., Ann. Med. 45 (2013) 446–54. doi:10.3109/07853890.2013.803739.
603
[31] V. Singhal, E.A. Lawson, K.E. Ackerman, P.K. Fazeli, H. Clarke, H. Lee, K. Eddy, D.A.
604
Marengi, N.P. Derrico, M.L. Bouxsein, M. Misra, Irisin levels are lower in young amenorrheic 605
athletes compared with eumenorrheic athletes and non-athletes and are associated with bone 606
density and strength estimates., PLoS One. 9 (2014) e100218.
607
doi:10.1371/journal.pone.0100218.
608
[32] G. Colaianni, C. Cuscito, T. Mongelli, P. Pignataro, C. Buccoliero, P. Liu, P. Lu, L. Sartini, 609
M. Di Comite, G. Mori, A. Di Benedetto, G. Brunetti, T. Yuen, L. Sun, J.E. Reseland, S.
610
Colucci, M.I. New, M. Zaidi, S. Cinti, M. Grano, The myokine irisin increases cortical bone 611
mass., Proc. Natl. Acad. Sci. U. S. A. 112 (2015) 12157–62. doi:10.1073/pnas.1516622112.
612
[33] A. Palermo, R. Strollo, E. Maddaloni, D. Tuccinardi, L. D’Onofrio, S.I. Briganti, G. Defeudis, 613
M. De Pascalis, M.C. Lazzaro, G. Colleluori, S. Manfrini, P. Pozzilli, N. Napoli, Irisin is 614
associated with osteoporotic fractures independently of bone mineral density, body 615
composition or daily physical activity., Clin. Endocrinol. (Oxf). 82 (2015) 615–9.
616
doi:10.1111/cen.12672.
617
[34] J.M. Kindler, N.K. Pollock, E.M. Laing, N.T. Jenkins, A. Oshri, C. Isales, M. Hamrick, R.D.
618
Lewis, Insulin Resistance Negatively Influences the Muscle-Dependent IGF-1-Bone Mass 619
Relationship in Premenarcheal Girls, J. Clin. Endocrinol. Metab. 101 (2016) 199–205.
620
doi:10.1210/jc.2015-3451.
621
[35] J.M. Kindler, N.K. Pollock, E.M. Laing, A. Oshri, N.T. Jenkins, C.M. Isales, M.W. Hamrick, 622
K.-H. Ding, D.B. Hausman, G.P. McCabe, B.R. Martin, K.M. Hill Gallant, S.J. Warden, C.M.
623
Weaver, M. Peacock, R.D. Lewis, Insulin Resistance and the IGF-I-Cortical Bone Relationship 624
in Children Ages 9-13 Years, J. Bone Miner. Res. (2017). doi:10.1002/jbmr.3132.
625
[36] E.M. Clark, A.R. Ness, J.H. Tobias, Adipose tissue stimulates bone growth in prepubertal 626
children., J. Clin. Endocrinol. Metab. 91 (2006) 2534–41. doi:10.1210/jc.2006-0332.
627
[37] M.B. Leonard, J. Shults, B.A. Wilson, A.M. Tershakovec, B.S. Zemel, Obesity during 628
childhood and adolescence augments bone mass and bone dimensions., Am. J. Clin. Nutr. 80 629
(2004) 514–523.
630
21 [38] Z.A. Cole, N.C. Harvey, M. Kim, G. Ntani, S.M. Robinson, H.M. Inskip, K.M. Godfrey, C.
631
Cooper, E.M. Dennison, Increased fat mass is associated with increased bone size but reduced 632
volumetric density in pre pubertal children, Bone. 50 (2012) 562–567.
633
doi:10.1016/j.bone.2011.05.005.
634
[39] H.T. Viljakainen, M. Pekkinen, E. Saarnio, H. Karp, C. Lamberg-Allardt, O. Makitie, Dual 635
effect of adipose tissue on bone health during growth, Bone. 48 (2011) 212–217.
636
doi:10.1016/j.bone.2010.09.022.
637
[40] A. Viitasalo, D.E. Laaksonen, V. Lindi, A.-M. Eloranta, J. Jääskeläinen, T. Tompuri, S.
638
Väisänen, H.-M. Lakka, T.A. Lakka, Clustering of Metabolic Risk Factors Is Associated with 639
High-Normal Levels of Liver Enzymes Among 6- to 8-Year-Old Children: The PANIC Study, 640
Metab. Syndr. Relat. Disord. 10 (2012) 337–343. doi:10.1089/met.2012.0015.
641
[41] J. Upadhyay, O.M. Farr, C.S. Mantzoros, The role of leptin in regulating bone metabolism., 642
Metabolism. 64 (2015) 105–13. doi:10.1016/j.metabol.2014.10.021.
643
[42] K. Liu, P. Liu, R. Liu, X. Wu, M. Cai, Relationship between serum leptin levels and bone 644
mineral density: A systematic review and meta-analysis, Clin. Chim. Acta. 444 (2015) 260–
645
263. doi:10.1016/j.cca.2015.02.040.
646
[43] W.L. do Prado, A. de Piano, M. Lazaretti-Castro, M.T. de Mello, S.G. Stella, S. Tufik, C.M.O.
647
do Nascimento, L.M. Oyama, M.C. Lofrano, L. Tock, D.A. Caranti, A.R. Dâmaso, 648
Relationship between bone mineral density, leptin and insulin concentration in Brazilian obese 649
adolescents, J. Bone Miner. Metab. 27 (2009) 613–619. doi:10.1007/s00774-009-0082-6.
650
[44] A. Sayers, N.J. Timpson, N. Sattar, J. Deanfield, A.D. Hingorani, G. Davey-Smith, J.H.
651
Tobias, Adiponectin and its association with bone mass accrual in childhood., J. Bone Miner.
652
Res. 25 (2010) 2212–20. doi:10.1002/jbmr.116.
653
[45] A. Böttner, J. Kratzsch, G. Müller, T.M. Kapellen, S. Blüher, E. Keller, M. Blüher, W. Kiess, 654
Gender differences of adiponectin levels develop during the progression of puberty and are 655
related to serum androgen levels., J. Clin. Endocrinol. Metab. 89 (2004) 4053–61.
656
doi:10.1210/jc.2004-0303.
657
[46] R. Weiss, J. Dziura, T.S. Burgert, W.V. Tamborlane, S.E. Taksali, C.W. Yeckel, K. Allen, M.
658
Lopes, M. Savoye, J. Morrison, R.S. Sherwin, S. Caprio, Obesity and the metabolic syndrome 659
in children and adolescents, N. Engl. J. Med. 350 (2004). doi:10.1056/NEJMoa031049.
660
[47] N.K. Pollock, P.J. Bernard, K. Wenger, S. Misra, B.A. Gower, J.D. Allison, H. Zhu, C.L.
661
Davis, Lower bone mass in prepubertal overweight children with prediabetes., J. Bone Miner.
662
Res. 25 (2010) 2760–9. doi:10.1002/jbmr.184.
663
[48] N.K. Pollock, P.J. Bernard, B. Gutin, C.L. Davis, H. Zhu, Y. Dong, Adolescent obesity, bone 664
22 mass, and cardiometabolic risk factors., J. Pediatr. 158 (2011) 727–34.
665
doi:10.1016/j.jpeds.2010.11.052.
666
[49] D.A. Lawlor, N. Sattar, A. Sayers, J.H. Tobias, The association of fasting insulin, glucose, and 667
lipids with bone mass in adolescents: findings from a cross-sectional study., J. Clin.
668
Endocrinol. Metab. 97 (2012) 2068–76. doi:10.1210/jc.2011-2721.
669
[50] R. Lucas, E. Ramos, A. Oliveira, T. Monjardino, H. Barros, Low-grade systemic inflammation 670
and suboptimal bone mineral density throughout adolescence: a prospective study in girls., 671
Clin. Endocrinol. (Oxf). 77 (2012) 665–71. doi:10.1111/j.1365-2265.2012.04430.x.
672
[51] C.S. Tam, K. Clément, L.A. Baur, J. Tordjman, Obesity and low-grade inflammation: a 673
paediatric perspective, Obes. Rev. 11 (2010) 118–126. doi:10.1111/j.1467- 674
789X.2009.00674.x.
675
[52] H.A. Bischoff-Ferrari, E. Giovannucci, W.C. Willett, T. Dietrich, B. Dawson-Hughes, 676
Estimation of optimal serum concentrations of 25-hydroxyvitamin D for multiple health 677
outcomes, Am. J. Clin. Nutr. 84 (2006) 18–28.
678
[53] Dietary Reference Intakes for Calcium and Vitamin D, National Academies Press, 679
Washington, D.C., 2011. doi:10.17226/13050.
680
[54] C. Lamberg-Allardt, M. Brustad, H.E. Meyer, L. Steingrimsdottir, Vitamin D - a systematic 681
literature review for the 5th edition of the Nordic Nutrition Recommendations., Food Nutr.
682
Res. 57 (2013) 1–31. doi:10.3402/fnr.v57i0.22671.
683
[55] M.F. Holick, N.C. Binkley, H.A. Bischoff-Ferrari, C.M. Gordon, D.A. Hanley, R.P. Heaney, 684
M.H. Murad, C.M. Weaver, Evaluation, treatment, and prevention of vitamin D deficiency:
685
An endocrine society clinical practice guideline, J. Clin. Endocrinol. Metab. 96 (2011) 1911–
686
1930. doi:10.1210/jc.2011-0385.
687
[56] C. Braegger, C. Campoy, V. Colomb, T. Decsi, M. Domellöf, M. Fewtrell, I. Hojsak, W.
688
Mihatsch, C. Molgaard, R. Shamir, D. Turck, J. van Goudoever, Vitamin D in the healthy 689
European paediatric population., J. Pediatr. Gastroenterol. Nutr. 56 (2013) 692–701.
690
doi:10.1097/MPG.0b013e31828f3c05.
691
[57] P. Arundel, S.F. Ahmed, J. Allgrove, N.J. Bishop, C.P. Burren, B. Jacobs, M.Z. Mughal, A.C.
692
Offiah, N.J. Shaw, British Paediatric and Adolescent Bone Group’s position statement on 693
vitamin D deficiency, BMJ. 345 (2012) e8182–e8182. doi:10.1136/bmj.e8182.
694
[58] S. Vandewalle, Y. Taes, T. Fiers, K. Toye, E. Van Caenegem, J.-M. Kaufman, J. De Schepper, 695
Relation of adrenal-derived steroids with bone maturation, mineral density and geometry in 696
healthy prepubertal and early pubertal boys., Bone. 69 (2014) 39–46.
697
doi:10.1016/j.bone.2014.09.002.
698
23 [59] A.B. Sopher, A.M. Jean, S.K. Zwany, D.M. Winston, C.B. Pomeranz, J.J. Bell, D.J. McMahon, 699
A. Hassoun, I. Fennoy, S.E. Oberfield, Bone age advancement in prepubertal children with 700
obesity and premature adrenarche: possible potentiating factors., Obesity (Silver Spring). 19 701
(2011) 1259–64. doi:10.1038/oby.2010.305.
702
[60] P. Utriainen, J. Jääskeläinen, A. Saarinen, E. Vanninen, O. Mäkitie, R. Voutilainen, Body 703
composition and bone mineral density in children with premature adrenarche and the 704
association of LRP5 gene polymorphisms with bone mineral density., J. Clin. Endocrinol.
705
Metab. 94 (2009) 4144–51. doi:10.1210/jc.2009-0315.
706
[61] V. Saraff, W. Högler, Osteoporosis in children: Diagnosis and management, Eur. J.
707
Endocrinol. 173 (2015) R185–R197. doi:10.1530/EJE-14-0865.
708
[62] V.H. Sidoroff, M.K. Ylinen, L.M. Kröger, H.P.J. Kröger, M.O. Korppi, Inhaled corticosteroids 709
and bone mineral density at school age: a follow-up study after early childhood wheezing., 710
Pediatr. Pulmonol. 50 (2015) 1–7. doi:10.1002/ppul.22968.
711
[63] Y.K. Loke, D. Gilbert, M. Thavarajah, P. Blanco, A.M. Wilson, Bone mineral density and 712
fracture risk with long-term use of inhaled corticosteroids in patients with asthma: systematic 713
review and meta-analysis., BMJ Open. 5 (2015) e008554. doi:10.1136/bmjopen-2015-008554.
714
[64] L. Margulies, M. Horlick, J.C. Thornton, J. Wang, E. Ioannidou, S.B. Heymsfield, 715
Reproducibility of pediatric whole body bone and body composition measures by dual-energy 716
X-ray absorptiometry using the GE Lunar Prodigy, J Clin.Densitom. 8 (2005) 298–304.
717
doi:JCD:8:3:298 [pii].
718
[65] J.A. Shepherd, L. Wang, B. Fan, V. Gilsanz, H.J. Kalkwarf, J. Lappe, Y. Lu, T. Hangartner, 719
B.S. Zemel, M. Fredrick, S. Oberfield, K.K. Winer, Optimal monitoring time interval between 720
DXA measures in children, J. Bone Miner. Res. 26 (2011) 2745–2752. doi:10.1002/jbmr.473.
721
[66] N.J. Crabtree, A. Arabi, L.K. Bachrach, M. Fewtrell, G. El-Hajj Fuleihan, H.H. Kecskemethy, 722
M. Jaworski, C.M. Gordon, Dual-energy x-ray absorptiometry interpretation and reporting in 723
children and adolescents: The revised 2013 ISCD pediatric official positions, J. Clin.
724
Densitom. 17 (2014) 225–242. doi:10.1016/j.jocd.2014.01.003.
725
